NEPA emerges as key to Trump PA win
The upshot: According to the NYT/Siena, Trump is down by 10 points. The margin of error is between 4.1 and 4.6 points (a pretty big margin of error). Nonetheless, it looks like Trump is in big trouble.
But is he really? Before laying out why the Trump campaign says "no," let me lay out why I personally think he might not be. The short version: Northeastern Pennsylvania (NEPA).
Let's start with a regional breakdown. As Ben Forstate, a PA maps guy, notes on Twitter:
The Times also broke the state into ReGioNs!: Allegheny: Biden +34 (63/29) Eastern PA: Biden +1 (44/43) Northeast PA: Biden +13(!) (48-35) Philly: Biden +60 (75/15) Philly Burbs: Biden +18 (53/18) Rest of PA: Trump +20 (39/59).
Biden winning NEPA by 13? I kind of doubt this. And so, it seems, do others. Check out liberal Washington Post journo, Dave Weigel, who throws cold water on the poll for just this reason:
NEPA is the killer - a region (Allentown, Scranton, etc) that swung hard to Trump in 2016, where Rs are trying to flip two Dem House seats. If Trump loses by 2016 margins in suburbs but loses those gains in NEPA, he has no path.
The districts Weigel is referring to are those currently occupied by Rep. Susan Wild and Rep. Matt Cartwright.
According to 270towin, in 2016, Hillary Clinton only won Wild's district by a whole 1.1 point.
And Trump won Cartwright's by 9.5 points.
Biden ahead by 13 there now? Well, if he is, as Weigel points out, Trump's goose is truly cooked. But that would be a mighty, mighty big swing from 2016. Would both the campaign and the NRCC have this so wrong? It seems unlikely.
Of course, the Allegheny number also signals a big - probably implausibly big- swing for Biden. As National Journal editor Alex Clearfield tweets:
That Allegheny number is big for Biden. Clinton won the county 56-40, which was in line with Obama's wins. He's got the Mt. Lebanon vote and the towns along the Monongahela/Ohio.
So this poll is saying that when Clinton and Obama were up by 16 or so points in the region, Biden is up by 34.
Clinton was a truly terrible candidate and Obama did make his Pennsylvania "Bittergate" comments which probably cost him a lot of votes in this region.
However, 34 points seems... unlikely.
But again, as Weigel is right to note, NEPA is what we really want to be watching here.
The Trump campaign has smartly declined to dissect this poll too much and engage in too much arguing about it. Their pushback is a bit "meh" for me, but to be fair, they want to spend their time attacking Biden, not engaging in number-crunching over a poll whose results will probably be overtaken by some other poll in a matter of days.
Where the Trump campaign is right:
They're saying the polls were wrong in 2016 (very much true with regard to a bunch of 2016 state-level polls).
Per Trump re-election campaign communications director Tim Murtaugh, "A lot of people know of Joe Biden but not very many know about Joe Biden" (also true: People know Biden's name, but they don't know his record, and while he bleats on about being "Scranton Joe," he hasn't lived in Scranton for about 100 years now).
Again, per Murtaugh, "This has been an unusual campaign and there are still more than four months to go and that is plenty of time to define an opponent" (and... true! Four months is an eternity in politics).
Where I think they're wrong: The campaign also says Biden's support for the Green New Deal will hurt him in Pennsylvania. Small problem: Biden's version of the GND isn't AOC's version or Bernie Sanders' version (it's pretty pro-nuclear and pro-natural gas - for example, Biden does not support a fracking ban).
But I'm sure the campaign will keep hammering away at this in an effort to make it true.
If those NEPA numbers are wrong, and there's good reason to believe they are, Trump does - as Weigel notes - still have a pretty viable path in Pennsylvania, even with attrition among suburban voters and Allegheny folks.
Pennsylvania ain't over til it's over... and right now, as bad as this polling looks on spec, there's good reason to think it is very far from over.